DharmakIrti starts, as mentioned above, by denying literal
omniscience
for the Buddha.
Buddhist-Omniscience
The word itself remains the same, but its meaning is undergoing a transforma-
tion.
We should mention here that there is another work attributed to
Nagarjuna and extant only in Chinese, the MahtJprajfltJpdramit6padeSa- sii5lra (MPPS). There is mention of both sarvajiia and sarviikilrajiia in this work, but following Hikata we may presume that the presence of
the latter, as well as any distinction between these two terms, is prob- ably more properly attributed to Kum! rajlva rather than N! garjuna. Yet even in this text, although both terms are used, there is no clear distinction between them, with Buddha acclaimed as omniscient (sarva-
jiia andlor sarv6ktJrajfia). while such an attainment is denied to HInaylna practitioners. This text is probably earlier than the Pm-
jr. tJptJramitil Satra in 25,()()() Lines mentioned above. Omniscience/or Vasubandhu
Following this transitional phase, we next come to that landmark of Buddhist literature, the Abhidharma-kosa. Composed around the mid- dle of the 4th century, this work represents the culmination of earlier Buddhist practices into a sophisticated philosophical system. Dating from after the start of the Mahayana, the Kosa is one of the last great
? ? ? ? BUDDHIST OMNISCIENCE
works of early Buddhism to be composed in India. Its author Vasuban- dhu has the distinction17 of being one of the only Buddhist authors to achieve renown in both branches of Buddhist philosophy, the earlier Hfnayana as well as the later Mahayana. Within the KoSa itself, however, we find only scant references to the concept of omniscience, which here retains its early form as sarvajna. The discussions of omnis- cience occur within the framework of sectarian disputations among the several groups (traditionally given as eighteen) which had arisen by this time, ? ch with their own special ways of interpreting Buddhist doc- trine.
We start with an objection from a Vl1tsrputrTya standpoint. This sect held that there must be an underlyinz basis of continuity 10 the person, which they called the pudgafa. This was regarded by most other sects as virtually identical to the lUman, and the Vl1tsrputrrya assertion of the pudgoJa was the object of repeated scorn and ridicule. In this context the Vl1tsfputrTya claimed that the pudgafa must have more than a momen- tary existence in order for Buddha to know everything. A mere series of mental instants cannot possibly have such knowledge. Vasubandhu replied that Buddha's omniscience is to be considered as a potential
ralher than an actualized state, dependent upon his attention or voli- tion. Vasubandhu cited the example of a fire, which cannot burn C\? erything at once, but has the ability to burn everything sequentially. Since Buddha has the ability to know everything, he is acknowledged to be omniscient. Thus Buddha's knowledge was regarded as having the same essential structure as the knowledge of ordinary people.
In opposition to this was Ihe position of the Mahl1sI1'f/ghikas, who held that Buddha's knowledge was qualitalively different from that of ordinary people. In the view typical of other Hlnayana schools, con- sciousness existed moment by moment, Hickering on and off like a movie. By the time a second momeDI arises, the first moment is com- pletely gone. However, this made it difficult to establish causality operating over any space of time_ The Sorviistivadins, whose views Vasubandhu generally upholds in the KOSO,IS asserted the existence of
" Shared with Wittgenstein, whose own philosophical career embraces IWO distinct ,,,,,,, .
\I AI least in the verJeS. In the commentary he often seem? 10 ? ide more with the ScUlrlInlilcll$. as noted in S . Anacker. Seven Worlcs of Vasubandh. . , p. 11.
? 41
? NAUOHTON
more than one moment; indeed, they claimed that both past and future existed just as much as the present. Their view was, as outlined above, similar to that of the Milinda-paflha, where Buddha's omniscience,
functioning in much the same way a s ordinary knowledge, is dependent upon volition for its activity. It is not automatic. The MahlJslJrrtghikas and others claimed that Buddha's omniscience was automatic. It was not dependent upon volition or any other factor, but was simply a result of his extreme mental purily_ This assertion seems quite similar to that of the early Jainas mentioned above. The Sarvastivadjns main- tained that Buddha's omniscience occurred over a period of two in- stants, one to cognize all phenomena and the other to render that cog- nition accessible. MahiJsaf! lghikas (ejected this, and said that since consciousness itself is self-luminous, there is no need of a second mo- ment. This position was to become standard in later Mahayana discus- sions of this topic, and of the nature of the mind in general.
As mentioned above, Vasubandhu's career as a Buddhist philoso- pher spans the division between Hlnayana and Mahayana, and so we find further discussions of omniscience in his later works. Of particular interest in this connection is the Mahayana Satriilamkl1ra (MSA), which along with the Abhisamayalarrtkara (AA) belongs to what are known as the "five texts of Maitreya. " But when we examine the MSA,
we find that its use of omniscience is quite difef rent from that of the AA. , which by itself would cast some serious doubt on the traditional ascription of these two texts to the same author. Let us proceed to the relevant textual citations. For the sake of analysis, we may divide them into two groups, first those that mention sarvajria (sj), and second those that mention sarviiklJrajrio (saj). First we find Vasubandhu's commentary on MSA I. IS: "the path to omniscience [sj) is extremely difficult 10 penetrate. " Here is a typical Buddhist or Upanifadic view
of omniscience as a melaphor for enlightenment. Next is MSA IX. I-2:
"After countless hundreds of ordeals, after countless halVests of births, and after countless destruction of blocks over immeasurable periods of time, omniscience [saj] is attained, unspoiled by (the slightest) block,like a casket of jewels thrown open, it is celebrated as
Buddhahood. " And in the commentary: "Buddhahood is presented from the point of view of achievement,nature,and a simile . . . Its nature is the attainment of omniscience Isail unspoiled by any block. " Here again omniscience is explicitly idenlified with Buddhahood or
? ? BUDDHIST OMNISCIENCE
enlightenment, as the removal of au obstacles or blocks to knowledge. Although the word used is sarvOklJrojflatlJ, there is little to distinguish it from the previous sarvaj"a. For a final instance we may cite MSA Xl. 2: "In brief, the SUlfa, Vinaya, and Abhidharma have a fourfold etymology. When he knows them the sage obtains omniscience (saj). " And its commentary: "In brief, Sutra, Vinaya and Abhidharma are each fourfold in etymology; the bodhisattva who knows them obtains omniscience [sj). " Where the verse has sarvlikoraj"all1, Vasubandhu ex- plains this as survajliutO. This is the clearest evidence we have that he makes no significant distinction between these two key terms. And the meaning of omniscience is here reduced to a simple mastery of the tradi- lional divisions of Buddhist literature. There are other quotes we could bringouthere,buttheywouldnotafef<:tthemainpoint. Thedistinc- lion at the heart of the AA between t h r e e different kinds of omnis- cience is unknown to Vasubandhu (at least at the time when he wrote this MSA commentary), and seems to be ignored by whomever com- posed the verses as well. The cruciaJ middle term which links the sor-
vajna of the Hrnayanists with the san'liklJraj"a of the Buddha in the AA is the mlJrgaj"a of the bodhisattvas, but this term does not occur even once in the MSA or Vasubandhu's commentary. This is a strong indica- tion that the AA system was nOI formulated by the same person who was responsible for the MSA, and renders the traditional ascription of both of these texts to Mailreya extremely dubious. The designation of the "five Maitreya texts" is unknown in the earliest catalog of Tibetan Ifanslations from Sanskrit texts, which was compiled in 824. 19
In passing, we may also note thai this distinction between different types of omniscience is also unknownin the Ratnagotra-vibhoga, other- wise known as the UUara? tantra, which is another one of the texts ascribed by the Tibelans 10 Maitreya. This text has been translated into
English twice, first by Obermiller and more recently by Takasaki. Ober- miller's translation was done from the Tibetan, but in the interval the Sanskrit text has been published by E. H. Johnston and Z. Nakamura.
" ? Set: Hakamaya Noriaki, "Some ProblelliS Concerning the TransmiloSion and Ap- pr<)priation of Yopc4ra Buddhism in Tilxt" (in Japanese) in Jourlal l oj O,ienlal Science, Vol. 21, No. 2 and Yoshimura Shll i, The l)enkll,? ma, An Oldesl Calalogue oj lire Ti? lan Buddhi! il CllnaM for thil date as well as the Observation about Ihe ab5cnce or the Maitrcya texIS as such.
? 4)
? N . . U. . G H T O N
There are many references to Buddha's omniscience in this text as well, and the Sanskrit text reveals that the word used in this context is sarva- jna. We shall cite one example, ROY 11. 42; "The Omniscience [sarva- jnatval}l of the Divine is That which is called the state of the Buddha, The Ultimate, Highest NirvllJa, The Buddha's inconceivable introspec- tion. ":ZO Here again omniscience is ? aid to be functionally equivalent to Buddhahood, a spiritual kind of omniscience, which is not distin- guished from the word for omniscience which was in use in the PaJi scriptures. This is further evidence that the Tibetan ascription of these three texts (the AA, MSA, and ROV) to the same author is highly questionable. And insofar as sarvOkllrajna is not found in the ROV, it appears that the system of the AA was unknown to its author as
well.
Omniscience in Later Mahllyllna
Following Vasubandhu by a few centuries is the career of the Bud- dbist logician Dharmakrrti, whose discussion of omniscience takes place partly in response to criticism From non-Buddhist sources, prin? cipaUy that of Kum4rila, a AAmll,! ,3Qko. 1' 10 lioe with the famous salutation to the Buddha by Dignlga as "the embodiment of valid k n o w l e d g e , " D h a r m a k i r t i ' s p r mi a r y c o n c e r n i s 1 0 e s t a b l i s h t h e credibility of Buddha's teachings. Thus he,like most other Buddhists, is not concerned with literal omniscience, but with a metaphorical or
. , E. E. Obennill? r, trans. , "The Sublime Science of the Great Vehicle to Salva? ticn" in ACla Orienlalia (1931), p. 2S2; Takasaki JikidO A Study QII the Rat? IIagotravibhllga (UI/artl(antra), p. 326; Nakamura Zuirytl The Ralnagotrtlvibhilga? Mllhil)'llnottDrD/DII/ra? 5l1Slra, p . 167. I quote from Obermli ler here b<<ause t prefer his Engli. h. Takasakl's veulon Ls: "That which Is called BuddhahOOd Is the Omniscience of the Self? born, The highest NirvlL(la, and the inconceivable Arhatship, Which is realized thrOUgh seLF. intro. pection. "
" This discu$. 'lion has already reaived t? . e anemion of several schoLars, . uch as Salkari Mookerjee, "The Omniscient as the Founder of a Religion" in NaWl ND/andll M? hQvihQ,a Rtsearch Publica/ion, Vol. 11; E. A. Solomon, " The Problem of Omnsi - cience (s. arvaj/latva)" ni AdyDr Librory Bulletin, Vol. XXVI. Parts 1-2: and Kawasaki ShinjO "Proofs of the Existence of an Omniscient Person" (in Japanese) in Epistemo/otyand Logic, Lec/ures in MDhQyDna Buddhism, Vol. 9 . My presentation of this material simply consists of a recapitulation of their prior work.
? ? ? 44
? BUDDHIST OMNISCIENCE
spiritual type. However, first let us lay OUI Kumlrila's arguments as our pQrva-pok? a.
Kumlrila, being a Mlmll'lsl aka, holds that all spiritual knowledge is derived from the Vedas exclusively. Thus any claims for other sources will meet with his disapproval. He states that a belief in a n omniscient person is a mere superstition, not founded on or provable by any logical means. There is no means for knowing spiritual truth other than the Vedas. In the case of an expert chef, we may praise him by saying Ihal he is omniscient with regard to cookery, and Kumarila will not object. Similarly in Ihe case of philosophical systems which classify reality into a set number of categories. One who has studied those categories may feel that he knows everything, in a general way. But it is impossible that anyone could know literally everything. Even a single body has more atoms and particles in it than can be known by one per- son, not to speak of the entire universe. Furthermore, human sense organs are restricted to their particular objects. Claims of omniscience arc as absurd as claiming thaI one can hear with one's nose, or see with one's ears. A person, such as our chef above, may be an expert in his chosen field, bUI this hardly implies that he knows anything at all out-
side of his specialty. There are certain limits to human ability. Through strenuous training, a man may be able to jump to a height of seven or eight feet, but no matter how much he practices, he will never be able to jump a mile high. Spiritual truths are simply inaccessible to human cognition without the assistance of the Vedas. Thus the Buddhist con- tention that their teacher knew such truths is simply mistaken. Any om- niscient person would necessarily know a number of repellent and disgusling things, which any sane person would avoid. Claims of om- niscience are based on testimony from the individuals concerned or from their followers, but the mere assertion does not make it so. Moreover, several different teachers have claimed (or been claimed) to be omniscient, yet they are unable to agree among themselves. This in- dicates that none of them are truly omniscient. Only an omniscient per- son can verify another's claim to be omniscient. The wise person will re- main skeptical of all such claims. Omniscience, if it exists, is impercepti- ble, and cannot be proved by a syllogism, blXause there is no logical sign of omniscience which could establish its existence. Inference is
always dependent upon perception. Thus the two main types of corrlXl knowledge are ruled out. Nor is there anything similar to an omniscient
45
? NAUGHTON
person, so his existence cannot be proved by analogy. Even if there were an omniscient person, he would retain his omniscience only as long as he was dri ectly cognizing everything, which presumably would be some kind of meditative state. Upon emerging from such a trance, he would lose his omniscience. Some pious Buddhists declare that Buddha himself did not say anything. but his teaching proceeded automatically based on the needs of his individual followers, but this is quite unbelievable. The Buddha did indeed give spiritual instructions to his followers, but his teaching may be wrong. Further objections are attributed to Samala and Vajnata. Does an omniscient person know e\'erything simultaneously or successively? If the former, one cognition would contain a multitude of contradictory qualities, which is impossi- ble (or at least unprecedented) for human cognition. If the latter, it would take a very lona time to know everything in the universe, during which such a person would most likdy grow old and die. A knowledge of the general nature of all phenomena will nOI do, since it doesn't qualify as a knowledge of every panicular thing. Also, such a knowl- edge could be false, and even i f it were true, it would reduce particular diversity to an indistinauishable unity, in which there would be no
distinctions such as teacher and pupil, right and wrong, etc. It may be said that the omniscient cognition is a special case, inaccessible to or- dinary people, but this statement itself is unproven and merely begs the question.
The foregoing arguments are to be found in the Tatlvaso'r! gfaha, a Buddhist work of the eighth century, which reports Kumarila's views in this case fairly and accurately. However, insofar as the above asser- tions constitute a PQrvQ-PQk? Q, they are incomplete without their refutation.
DharmakIrti starts, as mentioned above, by denying literal omniscience for the Buddha. Kumarila's critique may have some force for Jainas, but not aaainst a more limited form of omni! ICiem::e. Claims that Buddha was omniscient mean that what he taught is verifiably correct, and more specifically that he is an expert when it comes to anaining liberation, nirvtlfJQ, or enlightenment. Knowledge of mundane details such as the number of insects in the world or the number of fish in the oceans is irrelevanl. 12 The power of his physical vi-
lj, One is reminded here or the mention by Rabelais or "Mataeotechny-the Home or Usdw Knowledge. "
? ? ? . .
? 8UDDHIST OMNISCIENCE
sion is beside the point. Even if he were able 10 see tiny objects miles away, this would give no indication of his abilities as a spiritual teacher. And if we make far-sightedness a criterion for a good teacher, let us all go learn from the eagle! Buddha was a perfect teacher, in that he knew all the details concerning liberation. He had not only the per- sonal experience of his own enlightenment to inform the content of his discourse, but also the compassion for the suffering of others to pro- vide his motivation. In this sense, then, he is the perfect teacher, pra. s. i cd as omniscienl. Liberation results from the elimination of igno- rance, so his teaching is accepted as authoritative. In his capacity as the supreme teacher, Buddha must also theoretically have access to mun- dane information as well, to be used in the contellt of teaching as the situation demands. And while it may not be possible to prove that such an omniscient person actually ellists. it is no more possible to prove that such an existence is itself impossible. DharmakTni held the tradi- tional Buddhist opinion that consciousness is luminous by nature, and thai the various defilements and impurities which diminish its capacity are not essential but adventitious. Thus when consciousness is purified of all defilements, ilS capacity for knowledge should become infinite.
DbIlrmaklrti'li pOllitioll Oil this issue is reeosni:tably similar to that of the Milinda-paflho, and constitutes what has become the mainstream
Buddhist interprelation of Buddha's omniscience.
Santaraksita and his commentator KamaldTla give what is probably the most famous discussion of this problemll in their TaUwzsa'! fgraha, as mentioned above. In addition to the previous siudies by Solomon, Mookerjee, and olhers, the entire Touvoso'! fgraha has been translated into English. rendering this discussion accessible to a much wider au- dience. Here we shall give only a few of the highlights of this fascinat- ing discussion. In general, Santarak? ita says that the question of the existence of an omniscient person is open to doubt, and cannot be conclusively proven either way. Thus he criticizes the Mrmtlf! lSoka, saying: "Ir you deduce the incapacity of other persons, in regard to a certain effect. rrom your own Cltample . . . Ihen you land yourselr in absurdities. " The crilerion of omniKience is stated as: "One is 10 be recognised as omniscient only when he has been found to satisfy all
" Solomon. p. 67: "The TDII_1fIlrQhD's IrClIlmcnl of the ron? t ofstlrwViffl il the be$t thll we find in the whole . atliC or lodi. n philoSOphicai lilerllure. ? ?
? ? ? NAUGHTON
tests and au reasons, and has been found to have the true knowledge of all things. " This test is rather stringent, and would require omniscience on the pan of the judge as well. Yet Santaraksita's main concern is identical to that of DharmakTrti: "Whal is primarily and directly understood by us is that there is a Person who knows the means of at- taining Heaven and Liberation; -but not this alone; it is also believed that there is a Person who also kno",s all things. " Thus Santarak,ita goes beyond DharmakIrti's position in postulating the existence of $Omeone with a Literal omniscience. or cuursc, Buddha's status as the perfect teacher is the main issue, and here we find thal lhe Buddhist doc- trines are accepted as valid only after they have been critically ana- Iyted. In this conlext the goldsmith quote occurs again. Thus Buddha's doctrines are said to withstand rational criticism, and for this reason Buddha is hailed as omniscient. It is otherwise in the case of teachers whose doctrines rely on their alleged omniscience. The quality of omnis- cience being non-perceptible (al least to non-omniscient people), its non-apprehension is no proof of its non-existence, it is at most an oc-ca sion for doubt. Yet since Buddhist doctrines are said to be established by logical proofs, omniscience i s not regarded as Ihe exclusive property Of $akyamuni. If Other teachers propound dOClrines which agree with what has been proven before, or can withstand logical analysis, San- tarak$ita is willing to acknowledge theri omniscience as well. Insofar as they understand the true nature of reality, they aTe Buddhas. Omnis- cience results from the elimination of all obscurations and hindrances to cognition, as in Jainism. In this 'lay omniscience is not caused as much as it si revealed or uncovered. Slntarak$ita says that an omnis- cient person perceives everything directly through his mind, which or- dinarily correlates the data from the >cnSts, operates the memory, and so forth. Thus his knowledge is taken to be similar to ordinary knowledge, only carried to its limit by the repealed practice of medita- tion. For SAntarak,ita as for DharmakIrti, consciousness is luminous by nature, and removing obscuratioos reveals its luminoSity and in- creases its potential all the way to omniscience. "As a matter of fact, there is no limit to the knowledge of man. " In asserting that a "single clear appearance of all things in a single cognition is quite possible," S4ntarak$ita blurs one of the major distinctions made by (or for) Silkyamuni in the KOl,lt;lokotth% ? Sutto. The assertion that a "single cognition comprehends all that is knowable" follows the position of
. .
? BUDDHIST OMNISCIENCE
the MahDsorrghikas. In fact, Santaraksita wants to have it both ways: " Whatever He wishes to know He comes to know it without fail; -such is His power, as He has shaken off all evil. He knows things either simultaneously or in succession, just as He wishes," ? thus contlating the two kinds of omniscience which were so carefully distinguished by Slkyamuni for King Pasenadi above.
The problem of omniscience was a perennial concern in later MahayAna philosophy, as evidenced in the works of JiiAn? rfmitra, Ratnaktrti, and Mok? Akaragupta, but their treatment of this issue adds very little to that of Kamal? T1a. Two kinds of omniscience are distinguished, first and more important a spiritual or figurative omnis- cience which is equated with dharmajiia, especially knowledge of the four truths, and only secondarily a literal kind of omniscience, which is much harder to prove, and receives only cursory treatment. The former is regarded as the natural result of developing one's mental function
through repeated meditation. Both Ratnaklrti and Mok$1karagupta use the same example, that of the image of a girl which clearly appears to her lover based on his intense passion for her. Substituting the Bud- dhist four truths for the girl gives us their idea of omniscience. Rat- naklrti defines this kind of figurative omniscience as the capability or a mental function to become distinct, following the example of his teacher Ji'lAn? rimitra, who defines it as the knowledge of what is to be avoided and what is to be obtained which functions clearly in a great variety of situations. 204 This idea is a long way from literal omniscience. Literal omniscience receives scant attention from Ratnaklrti, although he does make a perfunctory effort to prove it as well. Here we will give MOk$Akaragupta's proof: "Words asreeing with a proof and having a definite object presuppose directly or indirectly the knowledge intuiting that object, as e. g. the words 'fire bums'; The words 'all produced things are momentary' also agree with a proof and have definite ob- j? ts; (therefore, the words presuppose the knowledge intuiting all
" Quoted by Ralnaktrti in Bijhnemann, Du AllwUKllde Bllddhil. Eill lhweu IIlId Rine Probleme. Rillnilkfrli's SQrv? iIlils ddhl iibenetzi lind kommenliert. p. 12. Also _E. Steinkellner'sartieleinL. Lancaster. o:d. ? PrVifIJPlJ,romilIJlindRefiltedSysrtms. p . )&7 (quoting J/Ilndrtmitra); "Throullh the force of such train in, it i5possibletllat a kind of mentat function. which i5 characteriud by matters to be obtaino:d. to be avoid- ed, and their kinds. becomes distinctly manifest as referring t o the &reatest number of
totally dear (individual) tire-situtations. Only this we consider a. . ! omniscience. "
"
? ? ? NAUGHTON
things, i. e. there must be an all-knowing person. I" This so-called proof is hardly convincing, and its very weakness is an indication of the relative unimportance it was accorded by Buddhist philosophers.
Grandiose as the conception of literal omniscience is, it is not yet the last word in Buddhist philosophy. In later MahayAna,ll ideas of the Buddha become even more fabulous , such that Buddha's "eye, car, nose, tongue, body, and mental consciousnesses arc each omniscient. Thus, a single moment of any consciousness cognizes all phenomena. " ? 26 For example, according to this idea, Buddha's knee is able to hear,
see, smell, taste, and feel all phenomena in the universe all the time, which takes the concept of omniscience about as far as it can go. By this time Buddha has become a god, or the god above the gods, and his followers, conveniently overlooking his earlier warnings, quite happily describe him in terms that far surpass their own experience.
Summary
We find that the concept of omniscience in Buddhist philosophy gives us a way of understanding the development of Buddhism within India. Prom examining the issues connected with this term we can easily discern a tendency within Buddhism to exalt the abstract at the expense of the concrete. To begin with, Sakyamuni was noticeably un-
comfortable with any claims for omniscience, made about either him or others. His concern was much more matter-of-fact, dealing with or- dinary experience and statements that could be made on that basis alone. However, soon after his death, we find that statements were at- nibuted to him which begin to open the door for claims by later Bud- dhists to be following an omniscient teacher. This is the natural out- come of early Buddhists wishing 10 uphold Ihe superiority of their own leacher and his doctrine in the face of rival claims to omniscience which were quite specific. Yet throughout the early period, at least, Buddha's omniscience does not go beyond the nature of ordinary knowledge, although it does represent the development of such knowl- edge to its limit. With the development of the Mahayana, the figure
? 1> Shading over into Tantra, although Hopkins typically gives no specific source for the followni g quote.
" Jeffrey Hopkin? , Med;/Q/;O" 0" Emptines,s p. L20. lO
? BUDDHIST OMNISCIENCE
of the human Sakyamuni is lost, replaced by a divinized and cosmic Buddha who is vastly superior to all olher creatures. Even Buddha's omniscience must become a super-omniscience 10 be worthy of this exalted being. Bridging Ihe gap between such an abstractly conceived super-godlike figure and Ihe ordinary individual is the figure of Ihe bodhisauva. Vel throughout this developmenl, Buddhisl philosophers consistently maintain thai omniscience in Buddhism is nOI to be understood literally, or that al leasl Ihis is not the primary sense in which Buddha is said to be omnisdcnl. Buddha is olUlliscielll by virtue of his perfect knowledge of the melhods and techniques for spiritual liberation, which find expression in his leachings. That this is to be dislinguished from literal omniscience is also indicaled by Tucci, Conze, Suzuki, and other translators as well, who render "sarvajiiaUJ" not as "omniscience," although this is the obvious choice, but by the somewhat cumbersome "All-knowledge. " This may perhaps be due to their desire to avoid the theological overtones of "omniscience" within Christian cultures.
Vel the later distinction belween sarvajiia and sarvlikllrajiia can be a useful one for clasSifying Buddhist scriptures. Without knowing pre- cisely when the lauer term was first introduced, we can nevertheless notice which lexts make use of it and which do not. This is a task for another occasion, but here we can note that it has been useful to us in criticizing the Tibetan tradition of the five Maiueya texts. Since sarvlikllrajiia is an odd term found especially in the Prajiillpllramitll Surra in 25,000 Lines and the Abhisamaylllamkllra, we can use it to distinguish which texts are associated with them.
?
tion.
We should mention here that there is another work attributed to
Nagarjuna and extant only in Chinese, the MahtJprajfltJpdramit6padeSa- sii5lra (MPPS). There is mention of both sarvajiia and sarviikilrajiia in this work, but following Hikata we may presume that the presence of
the latter, as well as any distinction between these two terms, is prob- ably more properly attributed to Kum! rajlva rather than N! garjuna. Yet even in this text, although both terms are used, there is no clear distinction between them, with Buddha acclaimed as omniscient (sarva-
jiia andlor sarv6ktJrajfia). while such an attainment is denied to HInaylna practitioners. This text is probably earlier than the Pm-
jr. tJptJramitil Satra in 25,()()() Lines mentioned above. Omniscience/or Vasubandhu
Following this transitional phase, we next come to that landmark of Buddhist literature, the Abhidharma-kosa. Composed around the mid- dle of the 4th century, this work represents the culmination of earlier Buddhist practices into a sophisticated philosophical system. Dating from after the start of the Mahayana, the Kosa is one of the last great
? ? ? ? BUDDHIST OMNISCIENCE
works of early Buddhism to be composed in India. Its author Vasuban- dhu has the distinction17 of being one of the only Buddhist authors to achieve renown in both branches of Buddhist philosophy, the earlier Hfnayana as well as the later Mahayana. Within the KoSa itself, however, we find only scant references to the concept of omniscience, which here retains its early form as sarvajna. The discussions of omnis- cience occur within the framework of sectarian disputations among the several groups (traditionally given as eighteen) which had arisen by this time, ? ch with their own special ways of interpreting Buddhist doc- trine.
We start with an objection from a Vl1tsrputrTya standpoint. This sect held that there must be an underlyinz basis of continuity 10 the person, which they called the pudgafa. This was regarded by most other sects as virtually identical to the lUman, and the Vl1tsrputrrya assertion of the pudgoJa was the object of repeated scorn and ridicule. In this context the Vl1tsfputrTya claimed that the pudgafa must have more than a momen- tary existence in order for Buddha to know everything. A mere series of mental instants cannot possibly have such knowledge. Vasubandhu replied that Buddha's omniscience is to be considered as a potential
ralher than an actualized state, dependent upon his attention or voli- tion. Vasubandhu cited the example of a fire, which cannot burn C\? erything at once, but has the ability to burn everything sequentially. Since Buddha has the ability to know everything, he is acknowledged to be omniscient. Thus Buddha's knowledge was regarded as having the same essential structure as the knowledge of ordinary people.
In opposition to this was Ihe position of the Mahl1sI1'f/ghikas, who held that Buddha's knowledge was qualitalively different from that of ordinary people. In the view typical of other Hlnayana schools, con- sciousness existed moment by moment, Hickering on and off like a movie. By the time a second momeDI arises, the first moment is com- pletely gone. However, this made it difficult to establish causality operating over any space of time_ The Sorviistivadins, whose views Vasubandhu generally upholds in the KOSO,IS asserted the existence of
" Shared with Wittgenstein, whose own philosophical career embraces IWO distinct ,,,,,,, .
\I AI least in the verJeS. In the commentary he often seem? 10 ? ide more with the ScUlrlInlilcll$. as noted in S . Anacker. Seven Worlcs of Vasubandh. . , p. 11.
? 41
? NAUOHTON
more than one moment; indeed, they claimed that both past and future existed just as much as the present. Their view was, as outlined above, similar to that of the Milinda-paflha, where Buddha's omniscience,
functioning in much the same way a s ordinary knowledge, is dependent upon volition for its activity. It is not automatic. The MahlJslJrrtghikas and others claimed that Buddha's omniscience was automatic. It was not dependent upon volition or any other factor, but was simply a result of his extreme mental purily_ This assertion seems quite similar to that of the early Jainas mentioned above. The Sarvastivadjns main- tained that Buddha's omniscience occurred over a period of two in- stants, one to cognize all phenomena and the other to render that cog- nition accessible. MahiJsaf! lghikas (ejected this, and said that since consciousness itself is self-luminous, there is no need of a second mo- ment. This position was to become standard in later Mahayana discus- sions of this topic, and of the nature of the mind in general.
As mentioned above, Vasubandhu's career as a Buddhist philoso- pher spans the division between Hlnayana and Mahayana, and so we find further discussions of omniscience in his later works. Of particular interest in this connection is the Mahayana Satriilamkl1ra (MSA), which along with the Abhisamayalarrtkara (AA) belongs to what are known as the "five texts of Maitreya. " But when we examine the MSA,
we find that its use of omniscience is quite difef rent from that of the AA. , which by itself would cast some serious doubt on the traditional ascription of these two texts to the same author. Let us proceed to the relevant textual citations. For the sake of analysis, we may divide them into two groups, first those that mention sarvajria (sj), and second those that mention sarviiklJrajrio (saj). First we find Vasubandhu's commentary on MSA I. IS: "the path to omniscience [sj) is extremely difficult 10 penetrate. " Here is a typical Buddhist or Upanifadic view
of omniscience as a melaphor for enlightenment. Next is MSA IX. I-2:
"After countless hundreds of ordeals, after countless halVests of births, and after countless destruction of blocks over immeasurable periods of time, omniscience [saj] is attained, unspoiled by (the slightest) block,like a casket of jewels thrown open, it is celebrated as
Buddhahood. " And in the commentary: "Buddhahood is presented from the point of view of achievement,nature,and a simile . . . Its nature is the attainment of omniscience Isail unspoiled by any block. " Here again omniscience is explicitly idenlified with Buddhahood or
? ? BUDDHIST OMNISCIENCE
enlightenment, as the removal of au obstacles or blocks to knowledge. Although the word used is sarvOklJrojflatlJ, there is little to distinguish it from the previous sarvaj"a. For a final instance we may cite MSA Xl. 2: "In brief, the SUlfa, Vinaya, and Abhidharma have a fourfold etymology. When he knows them the sage obtains omniscience (saj). " And its commentary: "In brief, Sutra, Vinaya and Abhidharma are each fourfold in etymology; the bodhisattva who knows them obtains omniscience [sj). " Where the verse has sarvlikoraj"all1, Vasubandhu ex- plains this as survajliutO. This is the clearest evidence we have that he makes no significant distinction between these two key terms. And the meaning of omniscience is here reduced to a simple mastery of the tradi- lional divisions of Buddhist literature. There are other quotes we could bringouthere,buttheywouldnotafef<:tthemainpoint. Thedistinc- lion at the heart of the AA between t h r e e different kinds of omnis- cience is unknown to Vasubandhu (at least at the time when he wrote this MSA commentary), and seems to be ignored by whomever com- posed the verses as well. The cruciaJ middle term which links the sor-
vajna of the Hrnayanists with the san'liklJraj"a of the Buddha in the AA is the mlJrgaj"a of the bodhisattvas, but this term does not occur even once in the MSA or Vasubandhu's commentary. This is a strong indica- tion that the AA system was nOI formulated by the same person who was responsible for the MSA, and renders the traditional ascription of both of these texts to Mailreya extremely dubious. The designation of the "five Maitreya texts" is unknown in the earliest catalog of Tibetan Ifanslations from Sanskrit texts, which was compiled in 824. 19
In passing, we may also note thai this distinction between different types of omniscience is also unknownin the Ratnagotra-vibhoga, other- wise known as the UUara? tantra, which is another one of the texts ascribed by the Tibelans 10 Maitreya. This text has been translated into
English twice, first by Obermiller and more recently by Takasaki. Ober- miller's translation was done from the Tibetan, but in the interval the Sanskrit text has been published by E. H. Johnston and Z. Nakamura.
" ? Set: Hakamaya Noriaki, "Some ProblelliS Concerning the TransmiloSion and Ap- pr<)priation of Yopc4ra Buddhism in Tilxt" (in Japanese) in Jourlal l oj O,ienlal Science, Vol. 21, No. 2 and Yoshimura Shll i, The l)enkll,? ma, An Oldesl Calalogue oj lire Ti? lan Buddhi! il CllnaM for thil date as well as the Observation about Ihe ab5cnce or the Maitrcya texIS as such.
? 4)
? N . . U. . G H T O N
There are many references to Buddha's omniscience in this text as well, and the Sanskrit text reveals that the word used in this context is sarva- jna. We shall cite one example, ROY 11. 42; "The Omniscience [sarva- jnatval}l of the Divine is That which is called the state of the Buddha, The Ultimate, Highest NirvllJa, The Buddha's inconceivable introspec- tion. ":ZO Here again omniscience is ? aid to be functionally equivalent to Buddhahood, a spiritual kind of omniscience, which is not distin- guished from the word for omniscience which was in use in the PaJi scriptures. This is further evidence that the Tibetan ascription of these three texts (the AA, MSA, and ROV) to the same author is highly questionable. And insofar as sarvOkllrajna is not found in the ROV, it appears that the system of the AA was unknown to its author as
well.
Omniscience in Later Mahllyllna
Following Vasubandhu by a few centuries is the career of the Bud- dbist logician Dharmakrrti, whose discussion of omniscience takes place partly in response to criticism From non-Buddhist sources, prin? cipaUy that of Kum4rila, a AAmll,! ,3Qko. 1' 10 lioe with the famous salutation to the Buddha by Dignlga as "the embodiment of valid k n o w l e d g e , " D h a r m a k i r t i ' s p r mi a r y c o n c e r n i s 1 0 e s t a b l i s h t h e credibility of Buddha's teachings. Thus he,like most other Buddhists, is not concerned with literal omniscience, but with a metaphorical or
. , E. E. Obennill? r, trans. , "The Sublime Science of the Great Vehicle to Salva? ticn" in ACla Orienlalia (1931), p. 2S2; Takasaki JikidO A Study QII the Rat? IIagotravibhllga (UI/artl(antra), p. 326; Nakamura Zuirytl The Ralnagotrtlvibhilga? Mllhil)'llnottDrD/DII/ra? 5l1Slra, p . 167. I quote from Obermli ler here b<<ause t prefer his Engli. h. Takasakl's veulon Ls: "That which Is called BuddhahOOd Is the Omniscience of the Self? born, The highest NirvlL(la, and the inconceivable Arhatship, Which is realized thrOUgh seLF. intro. pection. "
" This discu$. 'lion has already reaived t? . e anemion of several schoLars, . uch as Salkari Mookerjee, "The Omniscient as the Founder of a Religion" in NaWl ND/andll M? hQvihQ,a Rtsearch Publica/ion, Vol. 11; E. A. Solomon, " The Problem of Omnsi - cience (s. arvaj/latva)" ni AdyDr Librory Bulletin, Vol. XXVI. Parts 1-2: and Kawasaki ShinjO "Proofs of the Existence of an Omniscient Person" (in Japanese) in Epistemo/otyand Logic, Lec/ures in MDhQyDna Buddhism, Vol. 9 . My presentation of this material simply consists of a recapitulation of their prior work.
? ? ? 44
? BUDDHIST OMNISCIENCE
spiritual type. However, first let us lay OUI Kumlrila's arguments as our pQrva-pok? a.
Kumlrila, being a Mlmll'lsl aka, holds that all spiritual knowledge is derived from the Vedas exclusively. Thus any claims for other sources will meet with his disapproval. He states that a belief in a n omniscient person is a mere superstition, not founded on or provable by any logical means. There is no means for knowing spiritual truth other than the Vedas. In the case of an expert chef, we may praise him by saying Ihal he is omniscient with regard to cookery, and Kumarila will not object. Similarly in Ihe case of philosophical systems which classify reality into a set number of categories. One who has studied those categories may feel that he knows everything, in a general way. But it is impossible that anyone could know literally everything. Even a single body has more atoms and particles in it than can be known by one per- son, not to speak of the entire universe. Furthermore, human sense organs are restricted to their particular objects. Claims of omniscience arc as absurd as claiming thaI one can hear with one's nose, or see with one's ears. A person, such as our chef above, may be an expert in his chosen field, bUI this hardly implies that he knows anything at all out-
side of his specialty. There are certain limits to human ability. Through strenuous training, a man may be able to jump to a height of seven or eight feet, but no matter how much he practices, he will never be able to jump a mile high. Spiritual truths are simply inaccessible to human cognition without the assistance of the Vedas. Thus the Buddhist con- tention that their teacher knew such truths is simply mistaken. Any om- niscient person would necessarily know a number of repellent and disgusling things, which any sane person would avoid. Claims of om- niscience are based on testimony from the individuals concerned or from their followers, but the mere assertion does not make it so. Moreover, several different teachers have claimed (or been claimed) to be omniscient, yet they are unable to agree among themselves. This in- dicates that none of them are truly omniscient. Only an omniscient per- son can verify another's claim to be omniscient. The wise person will re- main skeptical of all such claims. Omniscience, if it exists, is impercepti- ble, and cannot be proved by a syllogism, blXause there is no logical sign of omniscience which could establish its existence. Inference is
always dependent upon perception. Thus the two main types of corrlXl knowledge are ruled out. Nor is there anything similar to an omniscient
45
? NAUGHTON
person, so his existence cannot be proved by analogy. Even if there were an omniscient person, he would retain his omniscience only as long as he was dri ectly cognizing everything, which presumably would be some kind of meditative state. Upon emerging from such a trance, he would lose his omniscience. Some pious Buddhists declare that Buddha himself did not say anything. but his teaching proceeded automatically based on the needs of his individual followers, but this is quite unbelievable. The Buddha did indeed give spiritual instructions to his followers, but his teaching may be wrong. Further objections are attributed to Samala and Vajnata. Does an omniscient person know e\'erything simultaneously or successively? If the former, one cognition would contain a multitude of contradictory qualities, which is impossi- ble (or at least unprecedented) for human cognition. If the latter, it would take a very lona time to know everything in the universe, during which such a person would most likdy grow old and die. A knowledge of the general nature of all phenomena will nOI do, since it doesn't qualify as a knowledge of every panicular thing. Also, such a knowl- edge could be false, and even i f it were true, it would reduce particular diversity to an indistinauishable unity, in which there would be no
distinctions such as teacher and pupil, right and wrong, etc. It may be said that the omniscient cognition is a special case, inaccessible to or- dinary people, but this statement itself is unproven and merely begs the question.
The foregoing arguments are to be found in the Tatlvaso'r! gfaha, a Buddhist work of the eighth century, which reports Kumarila's views in this case fairly and accurately. However, insofar as the above asser- tions constitute a PQrvQ-PQk? Q, they are incomplete without their refutation.
DharmakIrti starts, as mentioned above, by denying literal omniscience for the Buddha. Kumarila's critique may have some force for Jainas, but not aaainst a more limited form of omni! ICiem::e. Claims that Buddha was omniscient mean that what he taught is verifiably correct, and more specifically that he is an expert when it comes to anaining liberation, nirvtlfJQ, or enlightenment. Knowledge of mundane details such as the number of insects in the world or the number of fish in the oceans is irrelevanl. 12 The power of his physical vi-
lj, One is reminded here or the mention by Rabelais or "Mataeotechny-the Home or Usdw Knowledge. "
? ? ? . .
? 8UDDHIST OMNISCIENCE
sion is beside the point. Even if he were able 10 see tiny objects miles away, this would give no indication of his abilities as a spiritual teacher. And if we make far-sightedness a criterion for a good teacher, let us all go learn from the eagle! Buddha was a perfect teacher, in that he knew all the details concerning liberation. He had not only the per- sonal experience of his own enlightenment to inform the content of his discourse, but also the compassion for the suffering of others to pro- vide his motivation. In this sense, then, he is the perfect teacher, pra. s. i cd as omniscienl. Liberation results from the elimination of igno- rance, so his teaching is accepted as authoritative. In his capacity as the supreme teacher, Buddha must also theoretically have access to mun- dane information as well, to be used in the contellt of teaching as the situation demands. And while it may not be possible to prove that such an omniscient person actually ellists. it is no more possible to prove that such an existence is itself impossible. DharmakTni held the tradi- tional Buddhist opinion that consciousness is luminous by nature, and thai the various defilements and impurities which diminish its capacity are not essential but adventitious. Thus when consciousness is purified of all defilements, ilS capacity for knowledge should become infinite.
DbIlrmaklrti'li pOllitioll Oil this issue is reeosni:tably similar to that of the Milinda-paflho, and constitutes what has become the mainstream
Buddhist interprelation of Buddha's omniscience.
Santaraksita and his commentator KamaldTla give what is probably the most famous discussion of this problemll in their TaUwzsa'! fgraha, as mentioned above. In addition to the previous siudies by Solomon, Mookerjee, and olhers, the entire Touvoso'! fgraha has been translated into English. rendering this discussion accessible to a much wider au- dience. Here we shall give only a few of the highlights of this fascinat- ing discussion. In general, Santarak? ita says that the question of the existence of an omniscient person is open to doubt, and cannot be conclusively proven either way. Thus he criticizes the Mrmtlf! lSoka, saying: "Ir you deduce the incapacity of other persons, in regard to a certain effect. rrom your own Cltample . . . Ihen you land yourselr in absurdities. " The crilerion of omniKience is stated as: "One is 10 be recognised as omniscient only when he has been found to satisfy all
" Solomon. p. 67: "The TDII_1fIlrQhD's IrClIlmcnl of the ron? t ofstlrwViffl il the be$t thll we find in the whole . atliC or lodi. n philoSOphicai lilerllure. ? ?
? ? ? NAUGHTON
tests and au reasons, and has been found to have the true knowledge of all things. " This test is rather stringent, and would require omniscience on the pan of the judge as well. Yet Santaraksita's main concern is identical to that of DharmakTrti: "Whal is primarily and directly understood by us is that there is a Person who knows the means of at- taining Heaven and Liberation; -but not this alone; it is also believed that there is a Person who also kno",s all things. " Thus Santarak,ita goes beyond DharmakIrti's position in postulating the existence of $Omeone with a Literal omniscience. or cuursc, Buddha's status as the perfect teacher is the main issue, and here we find thal lhe Buddhist doc- trines are accepted as valid only after they have been critically ana- Iyted. In this conlext the goldsmith quote occurs again. Thus Buddha's doctrines are said to withstand rational criticism, and for this reason Buddha is hailed as omniscient. It is otherwise in the case of teachers whose doctrines rely on their alleged omniscience. The quality of omnis- cience being non-perceptible (al least to non-omniscient people), its non-apprehension is no proof of its non-existence, it is at most an oc-ca sion for doubt. Yet since Buddhist doctrines are said to be established by logical proofs, omniscience i s not regarded as Ihe exclusive property Of $akyamuni. If Other teachers propound dOClrines which agree with what has been proven before, or can withstand logical analysis, San- tarak$ita is willing to acknowledge theri omniscience as well. Insofar as they understand the true nature of reality, they aTe Buddhas. Omnis- cience results from the elimination of all obscurations and hindrances to cognition, as in Jainism. In this 'lay omniscience is not caused as much as it si revealed or uncovered. Slntarak$ita says that an omnis- cient person perceives everything directly through his mind, which or- dinarily correlates the data from the >cnSts, operates the memory, and so forth. Thus his knowledge is taken to be similar to ordinary knowledge, only carried to its limit by the repealed practice of medita- tion. For SAntarak,ita as for DharmakIrti, consciousness is luminous by nature, and removing obscuratioos reveals its luminoSity and in- creases its potential all the way to omniscience. "As a matter of fact, there is no limit to the knowledge of man. " In asserting that a "single clear appearance of all things in a single cognition is quite possible," S4ntarak$ita blurs one of the major distinctions made by (or for) Silkyamuni in the KOl,lt;lokotth% ? Sutto. The assertion that a "single cognition comprehends all that is knowable" follows the position of
. .
? BUDDHIST OMNISCIENCE
the MahDsorrghikas. In fact, Santaraksita wants to have it both ways: " Whatever He wishes to know He comes to know it without fail; -such is His power, as He has shaken off all evil. He knows things either simultaneously or in succession, just as He wishes," ? thus contlating the two kinds of omniscience which were so carefully distinguished by Slkyamuni for King Pasenadi above.
The problem of omniscience was a perennial concern in later MahayAna philosophy, as evidenced in the works of JiiAn? rfmitra, Ratnaktrti, and Mok? Akaragupta, but their treatment of this issue adds very little to that of Kamal? T1a. Two kinds of omniscience are distinguished, first and more important a spiritual or figurative omnis- cience which is equated with dharmajiia, especially knowledge of the four truths, and only secondarily a literal kind of omniscience, which is much harder to prove, and receives only cursory treatment. The former is regarded as the natural result of developing one's mental function
through repeated meditation. Both Ratnaklrti and Mok$1karagupta use the same example, that of the image of a girl which clearly appears to her lover based on his intense passion for her. Substituting the Bud- dhist four truths for the girl gives us their idea of omniscience. Rat- naklrti defines this kind of figurative omniscience as the capability or a mental function to become distinct, following the example of his teacher Ji'lAn? rimitra, who defines it as the knowledge of what is to be avoided and what is to be obtained which functions clearly in a great variety of situations. 204 This idea is a long way from literal omniscience. Literal omniscience receives scant attention from Ratnaklrti, although he does make a perfunctory effort to prove it as well. Here we will give MOk$Akaragupta's proof: "Words asreeing with a proof and having a definite object presuppose directly or indirectly the knowledge intuiting that object, as e. g. the words 'fire bums'; The words 'all produced things are momentary' also agree with a proof and have definite ob- j? ts; (therefore, the words presuppose the knowledge intuiting all
" Quoted by Ralnaktrti in Bijhnemann, Du AllwUKllde Bllddhil. Eill lhweu IIlId Rine Probleme. Rillnilkfrli's SQrv? iIlils ddhl iibenetzi lind kommenliert. p. 12. Also _E. Steinkellner'sartieleinL. Lancaster. o:d. ? PrVifIJPlJ,romilIJlindRefiltedSysrtms. p . )&7 (quoting J/Ilndrtmitra); "Throullh the force of such train in, it i5possibletllat a kind of mentat function. which i5 characteriud by matters to be obtaino:d. to be avoid- ed, and their kinds. becomes distinctly manifest as referring t o the &reatest number of
totally dear (individual) tire-situtations. Only this we consider a. . ! omniscience. "
"
? ? ? NAUGHTON
things, i. e. there must be an all-knowing person. I" This so-called proof is hardly convincing, and its very weakness is an indication of the relative unimportance it was accorded by Buddhist philosophers.
Grandiose as the conception of literal omniscience is, it is not yet the last word in Buddhist philosophy. In later MahayAna,ll ideas of the Buddha become even more fabulous , such that Buddha's "eye, car, nose, tongue, body, and mental consciousnesses arc each omniscient. Thus, a single moment of any consciousness cognizes all phenomena. " ? 26 For example, according to this idea, Buddha's knee is able to hear,
see, smell, taste, and feel all phenomena in the universe all the time, which takes the concept of omniscience about as far as it can go. By this time Buddha has become a god, or the god above the gods, and his followers, conveniently overlooking his earlier warnings, quite happily describe him in terms that far surpass their own experience.
Summary
We find that the concept of omniscience in Buddhist philosophy gives us a way of understanding the development of Buddhism within India. Prom examining the issues connected with this term we can easily discern a tendency within Buddhism to exalt the abstract at the expense of the concrete. To begin with, Sakyamuni was noticeably un-
comfortable with any claims for omniscience, made about either him or others. His concern was much more matter-of-fact, dealing with or- dinary experience and statements that could be made on that basis alone. However, soon after his death, we find that statements were at- nibuted to him which begin to open the door for claims by later Bud- dhists to be following an omniscient teacher. This is the natural out- come of early Buddhists wishing 10 uphold Ihe superiority of their own leacher and his doctrine in the face of rival claims to omniscience which were quite specific. Yet throughout the early period, at least, Buddha's omniscience does not go beyond the nature of ordinary knowledge, although it does represent the development of such knowl- edge to its limit. With the development of the Mahayana, the figure
? 1> Shading over into Tantra, although Hopkins typically gives no specific source for the followni g quote.
" Jeffrey Hopkin? , Med;/Q/;O" 0" Emptines,s p. L20. lO
? BUDDHIST OMNISCIENCE
of the human Sakyamuni is lost, replaced by a divinized and cosmic Buddha who is vastly superior to all olher creatures. Even Buddha's omniscience must become a super-omniscience 10 be worthy of this exalted being. Bridging Ihe gap between such an abstractly conceived super-godlike figure and Ihe ordinary individual is the figure of Ihe bodhisauva. Vel throughout this developmenl, Buddhisl philosophers consistently maintain thai omniscience in Buddhism is nOI to be understood literally, or that al leasl Ihis is not the primary sense in which Buddha is said to be omnisdcnl. Buddha is olUlliscielll by virtue of his perfect knowledge of the melhods and techniques for spiritual liberation, which find expression in his leachings. That this is to be dislinguished from literal omniscience is also indicaled by Tucci, Conze, Suzuki, and other translators as well, who render "sarvajiiaUJ" not as "omniscience," although this is the obvious choice, but by the somewhat cumbersome "All-knowledge. " This may perhaps be due to their desire to avoid the theological overtones of "omniscience" within Christian cultures.
Vel the later distinction belween sarvajiia and sarvlikllrajiia can be a useful one for clasSifying Buddhist scriptures. Without knowing pre- cisely when the lauer term was first introduced, we can nevertheless notice which lexts make use of it and which do not. This is a task for another occasion, but here we can note that it has been useful to us in criticizing the Tibetan tradition of the five Maiueya texts. Since sarvlikllrajiia is an odd term found especially in the Prajiillpllramitll Surra in 25,000 Lines and the Abhisamaylllamkllra, we can use it to distinguish which texts are associated with them.
?
